June 2005 - Glasgow-haskell-users

Map, Set libraries
by Serge D. Mechveliani 03 Jun '05

03 Jun '05

Thanks to people for the explanations. To my question >> 1. elems :: Set a -> [a] >> setToList :: Set a -> [a] >> >> These two look like synonyms, but have different comments. >> Am I missing something? Jens Fisseler <jens.fisseler(a)fernuni-hagen.de> writes on 2 Jun 2005 > Both functions compute the same list, and IMHO the comments state the > same. I meant the description in the ghc-6.4 documentation. Data.Set.html says " elems :: Set a -> [a] O(n). The elements of a set. setToList :: Set a -> [a] O(n). Convert the set to a list elements. " What is the difference? If they are really equivalent, then, it is natural either to remove `elems' or to follow the general GHC policy and to move setToList to the `Obsolete' list section in documentation, like it was done with many other functions for Set and FiniteMap. - ? ----------------- Serge Mechveliani mechvel(a)botik.ru

5 7

addListToFM
by Serge D. Mechveliani 03 Jun '05

03 Jun '05

I used to apply FiniteMap.addListToFM. Now, what is its best expression in ghc-6.4 ? Is this \ mp pairs -> Map.union (Map.fromList pairs) mp ? (the order of the arguments in Map.union is essential). Thank you in advance for explanation. ----------------- Serge Mechveliani mechvel(a)botik.ru

3 2

Map, Set
by Serge D. Mechveliani 03 Jun '05

03 Jun '05

On 2 Jun 2005 S. Alexander Jacobson <alex(a)alexjacobson.com> writes > Any reason the libaries don't define: > > class HasNull a where null::a->Bool > class HasEmpty a where empty::a > > I find that I sometimes switch between using lists, sets, or tables as > my collection type and the forced import qualifification for generic > collection operations seems annoying. Probably, it would be natural for the Map and Set GHC libraries to introduce some version of class SetLike where ... My programs also switch with similar operations between sets and various tables. To unify the interface, I defined the following: ------------------------------------------------------------------- class Cardinality a where card :: a -> Natural -- (Integer) instance Cardinality (Set.Set a) where card = genericLength . Set.elems instance Cardinality (Map.Map k a) where card = genericLength . Map.keys class Cardinality a => SetLike a where sEmpty :: a sIsEmpty :: a -> Bool sUnion :: [a] -> a sMinus :: a -> a -> a sIntersect :: [a] -> a sConcat :: [a] -> a sNub :: a -> a instance Ord a => SetLike (Set.Set a) where sEmpty = Set.empty sIsEmpty = Set.null sUnion = Set.unions sMinus = Set.difference sNub = id sConcat = sUnion sIntersect [] = error "sIntersect []\n" sIntersect (s:ss) = foldl Set.intersection s ss sIncluded :: SetLike a => a -> a -> Bool sIncluded s = sIsEmpty . sMinus s instance SetLike IdList where ... ------------------------------------------------------------------- This was as under ghc-6.2.2, and now it is edited to meet 6.4. About what Set and Map libraries could provide ---------------------------------------------- The candidates may be `any', `all' analogues for Set, Map (anyway, they are simply expressed by composing with toList). direct product of sets (can be simply expressed by composing with toList, fromList). composition of maps, inverse map. On standard ----------- It is very desirable to put the Map and Set libraries to future standard Haskell library. With this, our struggle with their interface changes will end. I wonder how the Haskell language and library are going on, whether Haskell-2 is coming. Regards, ----------------- Serge Mechveliani mechvel(a)botik.ru

2 1

[ ghc-Feature Requests-1212959 ] Reading source files in encodings other than Latin-1
by SourceForge.net 03 Jun '05

03 Jun '05

Feature Requests item #1212959, was opened at 2005-06-01 18:53 Message generated for change (Comment added) made by simonmar You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=358032&aid=1212959&group_… Please note that this message will contain a full copy of the comment thread, including the initial issue submission, for this request, not just the latest update. >Category: None >Group: None Status: Open Priority: 5 Submitted By: Nobody/Anonymous (nobody) Assigned to: Nobody/Anonymous (nobody) >Summary: Reading source files in encodings other than Latin-1 Initial Comment: For GHC 6.4 on SuSE Linux 9.2 (installed from the GHC RPM). When including SOME non-ascii characters in character or string literals, in the source code, I get a "lexical error in string/character literal" error message. Examples are some french accents and german umlauts - é,Ü,Ä. However, with other characters from the same set (like üä) there is no problem, they are processed correctly I checked with Hugs and there were no problems, so it seems to be a GHC bug. For further questions send email to: Rainer Volz, mail at vrtprj.com ---------------------------------------------------------------------- >Comment By: Simon Marlow (simonmar) Date: 2005-06-03 08:29 Message: Logged In: YES user_id=48280 I don't know about wxHaskell, it's possible that it is using your system's default encoding. Thanks for the report anyway. I'll re-brand it as a feature request. ---------------------------------------------------------------------- Comment By: Nobody/Anonymous (nobody) Date: 2005-06-02 16:24 Message: Logged In: NO Ah ok, I read so much about Haskell and Unicode that I didn't think about that. My system uses UTF-8 as standard encoding, so Emacs saves the source file also in that encoding. Using ISO-8859-1 as encoding to save the file helps with the lexical error. However the strings are still not displayed correctly. I tried it with texts in wxhaskell windows and with simple putStr "Üo" scripts. Do I have to change my system's encoding to IOS8859-1 to get proper results? ---------------------------------------------------------------------- Comment By: Simon Marlow (simonmar) Date: 2005-06-02 08:32 Message: Logged In: YES user_id=48280 It works fine for me. What encoding are you using? GHC only understands the Latin-1 (ISO8859-1) encoding for source files. If you are using Latin-1, then please attach a source file that we can test. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=358032&aid=1212959&group_…

1 0

doc on Map.insert
by Serge D. Mechveliani 02 Jun '05

02 Jun '05

The ghc-6.4 documentation says about Data.Map " insert :: Ord k => k -> a -> Map k a -> Map k a O(log n). Insert a new key and value in the map. " If this is to replace old addToFM, then it is better to remove the word "new". Right? ----------------- Serge Mechveliani mechvel(a)botik.ru

1 0

RE: Re[2]: [Haskell-cafe] ghc 6.4 import problem
by Simon Marlow 02 Jun '05

02 Jun '05

On 01 June 2005 07:24, Bulat Ziganshin wrote: > Hello Fergus, > > Tuesday, May 31, 2005, 10:24:41 PM, you wrote: > >>> import Data.Set hiding (map) >>> import qualified Data.Set as Set >>> >>> will do fine. > >> That code only compiles with ghc 6.4, and won't compile with ghc 6.2: >> you'll get an error for the "hiding (map)" part, because in 6.2 >> Data.Set does not contain a "map" function. > > btw, question to Simon Marlow and Simon PJ - how about disabling this > error message? You mean allowing import M hiding (x) even if x is not exported by M? Or making it a warning? That's certainly possible, but it constitutes an extension to Haskell 98, so it would at least require the -fglasgow-exts flag. I think you're better off keeping your code portable. Cheers, Simon

2 1

RE: .depend in base package
by Simon Marlow 02 Jun '05

02 Jun '05

On 02 June 2005 08:22, Krasimir Angelov wrote: > I just tried to update and build my CVS tree, but the compilation > failed for base package. In the 'make boot' the .depend file is > created but it contains only the C part. The compiler is invoked as > normal and there isn't any error. The compilation is failing latter > when it tries to build Control.Arrow before Prelude. Do you have any > idea what is going wrong? It was working when I tried it last time. Fixed now, sorry about that. Cheers, Simon

1 0

RE: make html
by Simon Marlow 02 Jun '05

02 Jun '05

On 01 June 2005 12:51, Serge D. Mechveliani wrote: > The problem was that I could not find these .gz archives by > following the www references. I wonder what I was missing. > Now, as I read your letter, I go to .../docs/latest/ and observe > them. The printable documentation is linked from the documentation page: http://www.haskell.org/ghc/documentation.html. Cheers, Simon

2 1

RE: Simplifier output explanation needed
by Simon Marlow 02 Jun '05

02 Jun '05

On 01 June 2005 13:54, Jens Fisseler wrote: > I'm currently trying to optimize some code, in particular trying to > make some functions as strict as possible. To do this, I quite often > look for strictness annotations either in the interface files or the > the simplifier output. Doing this, something odd occured to me: > inserting a '$!' into a function call, trying to enforce eager > evaluation, the simplifier output changed. Ok, I should expect this, > but I don't understand the change. The function name got '$s' as a > prefix and the strictness annotation ('Str: DmdType SSL') vanished. > > So my question is what all those different prefixes mean. I've > stumbled across at least '$s' and '$w' (worker wrapper?). Why did the > strictness annotation disappear? Is there any documentation that can > enlighten me? $s means "specialised" I believe (Simon PJ is away until next week, so you'll have to wait until then for a definitive answer). Cheers, Simon

1 0

.depend in base package
by Krasimir Angelov 02 Jun '05

02 Jun '05

Hello, Guys I just tried to update and build my CVS tree, but the compilation failed for base package. In the 'make boot' the .depend file is created but it contains only the C part. The compiler is invoked as normal and there isn't any error. The compilation is failing latter when it tries to build Control.Arrow before Prelude. Do you have any idea what is going wrong? It was working when I tried it last time. Cheers, Krasimir # DO NOT DELETE: Beginning of C dependencies System/CPUTime_hsc.o System/CPUTime_hsc.p_o: System/CPUTime_hsc.c System/Time_hsc.o System/Time_hsc.p_o: System/Time_hsc.c Text/Regex/Posix_hsc.o Text/Regex/Posix_hsc.p_o: Text/Regex/Posix_hsc.c # DO NOT DELETE: End of C dependencies # DO NOT DELETE: Beginning of Haskell dependencies # DO NOT DELETE: End of Haskell dependencies

1 0